基于密文KNN检索的室内定位隐私保护算法

Indoor Location Privacy Protection Algorithm Based on Ciphertext KNN Retrieval

在定位请求服务中,如何保护用户的位置隐私和位置服务提供商(Localization service provider,LSP)的数据隐私是关系到WiFi指纹定位应用的一个具有挑战性的问题。基于密文域的K-近邻(K-nearest neighbors,KNN)检索,本文提出了一种适用于三方的定位隐私保护算法,能有效提升对LSP指纹信息隐私的保护强度并降低计算开销。服务器和用户分别完成对指纹信息和定位请求的加密,而第三方则基于加密指纹库和加密定位请求,在隐私状态下完成对用户的位置估计。所提算法把各参考点的位置信息随机嵌入指纹,可避免恶意用户获取各参考点的具体位置;进一步利用布隆滤波器在隐藏接入点信息的情况下,第三方可完成参考点的在线匹配,实现对用户隐私状态下的粗定位,可与定位算法结合降低计算开销。在公共数据集和实验室数据集中,对两种算法的安全、开销和定位性能进行了全面的评估。与同类加密算法比较,在不降低定位精度的情况下,进一步增强了对数据隐私的保护。

In the location request service,how to protect the user’s location privacy and the data privacy of the location service provider(LSP)is a challenging issue related to WiFi fingerprinting applications.Based on the K-nearest neighbors(KNN)retrieval of the ciphertext,this paper proposes a positioning privacy protection algorithm suitable for the three party,which can effectively improve the protection intensity of the privacy of LSP fingerprint information and reduce calculation overhead.The positioning algorithm is completed by a third party based on the encrypted fingerprint database and encrypted positioning request,which is completed in the state of privacy.Through the random embedding of the location information in the fingerprint,the algorithm can avoid the physical location of the reference point(RP)in the fingerprint database.The Bloom filter(BF)is further used to complete the online matching of the reference point when hiding the access point information,which achieves rough positioning in the privacy of the user,and reduces the calculation overhead with the positioning algorithm.In the data set of public datasets and laboratory data,the security,expense and positioning performance of the two algorithms have been comprehensively evaluated.Compared with similar encryption algorithms,without reducing positioning accuracy,it further enhances the protection of data privacy.